Lat. Am. J. Aquat. Res., 42(3): 604-621, 2014 Massive salp outbreaks in the inner sea of Chiloé Island 604 1 DOI: 103856/vol42-issue3-fulltext-18

Research Article

Massive salp outbreaks in the inner sea of Chiloé Island (Southern ): possible causes and ecological consequences

Ricardo Giesecke1,2, Alejandro Clement3, José Garcés-Vargas1, Jorge I. Mardones4 Humberto E. González1,6, Luciano Caputo1,2 & Leonardo Castro5,6 1Instituto de Ciencias Marinas y Limnológicas, Facultad de Ciencias, Universidad Austral de Chile P.O. Box 567, , Chile 2Centro de Estudios en Ecología y Limnología Chile, Geolimnos, Carelmapu 1 N°540, Valdivia, Chile 3Plancton Andino, P.O. Box 823, , Chile 4Institute for Marine and Antarctic Studies, University of Tasmania Private Bag 55, Hobart, Tasmania 7001, 5Departamento de Oceanografía, Universidad de Concepción, P.O. Box 160-C, Concepción, Chile 6Programa de Financiamiento Basal, COPAS Sur-Austral y Centro COPAS de Oceanografía Universidad de Concepción, P.O. Box 160-C, Concepción, Chile

ABSTRACT. During 2010 several massive salp outbreaks of the Subantarctic species Ihlea magalhanica were recorded in the inner sea of Chiloé Island (ISCh, Southern Chile), affecting both phytoplankton abundance and salmon farmers by causing high fish mortality. First outbreaks were recorded during February 2010 when Ihlea magalhanica reached up to 654,000 ind m-3 close to the net pens in Maillen Island and consecutive outbreaks could be followed during March and from October to November 2010. One month prior to the first recorded salp outbreak, the adjacent oceanic region and ISCh showed a sharp decline of ca. 1.0ºC in sea surface temperature and an atypical pattern of oceanic sea surface currents, changing from a predominantly meridional (northward) to a zonal (eastward) direction, probably causing a massive Subantarctic Water parcel to enter the ISCh. During the outbreaks, surface chlorophyll concentration decreased from an historical mean of 13.8 to less than 4 mg Chl-a m-3, and did not return to normal conditions throughout the entire year, and similar results were also observed in phytoplankton abundance. The abundance of salp aggregations were highest close to the salmon net pens, which acted as physical barriers, and may have favored the successful reproduction and persistence of the outbreaks during 2010. The possible impact of these outbreaks on phytoplankton quality and quantity, as well as potential scenarios for the development of further outbreaks is discussed. Keywords: Ihlea magalhanica, outbreak, salmon mortality, Chiloé Island, southern Chile.

Proliferaciones masivas de salpas en el mar interior de la isla de Chiloé (sur de Chile): posibles causas y consecuencias ecológicas

RESUMEN. Durante 2010 varias proliferaciones masivas de salpa subantártica, Ihlea magalhanica, fueron registrados en el mar interior de Chiloé (MICh, sur de Chile ), que afectó la abundancia de fitoplancton y a los productores de salmón al causar altas mortalidades de peces. Las primeras proliferaciones se registraron en febrero de 2010, cuando la abundancia de I. magalhanica alcanzó 654.000 ind m-3 cerca de las balsas jaulas en isla Maillen. Proliferaciones consecutivas fueron posteriormente registradas durante marzo y de octubre a noviembre de 2010. Un mes antes de los registros de las primeras proliferaciones, la región oceánica adyacente y el MICh evidenciaron un fuerte descenso en la temperatura superficial del mar de ca. 1ºC y un patrón atípico de las corrientes oceánicas superficiales, cambiando desde una circulación predominantemente meridional (norte) a zonal (hacia el este), probablemente causando el ingreso de una parcela de agua subantártica al mar interior de Chiloé. Durante las proliferaciones, la concentración de clorofila (Clor-a) superficial disminuyó de una media histórica de 13,8 a menos de 4 mg Clor-a m-3, manteniéndose bajo los promedios históricos durante todo año 2010. Resultados similares se observaron también en la abundancia de fitoplancton. La abundancia de las concentraciones de salpas fueron mayores cerca de las balsas jaulas de salmón, que actúan como barreras físicas, concentrando los organismos y posiblemente favoreciendo el éxito reproductivo y la persistencia de estas proliferaciones durante el 2010. Se discute el posible impacto de estas proliferaciones en la calidad y cantidad de fitoplancton, así como los posibles escenarios para el desarrollo de nuevos eventos. 2605 Latin American Journal of Aquatic Research

Palabras clave: Ihlea magalhanica, proliferaciones, mortalidad de salmones, isla de Chiloé, sur de Chile. ______Corresponding author: Ricardo Giesecke ([email protected])

INTRODUCTION favorable (Goy et al., 1989; CIESM, 2001; Molinero et al., 2005). Salps are capable of filtering particles even The sudden increase of massive gelatinous zooplankton in the 0.1 to 1 µm range, making them able to feed upon outbreaks in several coastal areas around the world has small bacteria, Prochlorococcus and colloids (Sutherland become more frequent, especially during the last few et al., 2010), thereafter inducing an exceptional food decades. The most documented case is the intrusion of web shortcut between small particles and higher trophic the non-indigenous ctenophore species Mnemiopsis levels, bypassing the microbial loop. The wide size leidyi in the Black Sea during the 80’s, which latter range of potential food sources, high clearance rates spread into the Caspian, Baltic, and North Seas (250-801 mL mg-1 h-1) (Deibel, 1982a; Mullin, 1983), (Shiganova et al., 2001; Faasse & Bayha, 2006; and reproductive cycle strategy allows them to achieve Javidpour et al., 2006), considerably affecting the food exceptional growth rates, ranging between 0.3% and web structure and possibly helping to promote the 28% body length per hour (Deibel, 1982b; Le Borgne collapse of fisheries in the Black Sea (Kideys, 1994, & Moll, 1986). Each solitary individual (oozooid) 2002). Major gelatinous outbreaks were also reported produces, by budding, a stolon which strobilates into in 2011 and serious problems in nuclear plants in Israel, chains, between 25 and 75 aggregated individuals Japan and USA were reported where the seawater (blastozooids), depending on the oozooid size (Heron cooling systems became clogged with jellyfish (Purcell, & Benham, 1985), for Thalia democratica. These 2012). In addition, the bursting of fishing nets caused chains are fertilized shortly after they detach from the by the giant Nemopilema nomurai in the coasts of Japan stolon to form a free living, pseudo-colonial group. The have forced fishermen to adapt fishing equipment in latter process allows salps to achieve high densities in order to keep out jelly swarms (Uye, 2008). short periods of time, while mortality is likely minimal The impacts of these gelatinous zooplankton as a result of viviparity (Heron, 1972), and causes the outbreaks on tourism have also increased considerably overgrazing of food resources during these short burst and include the temporary closing of beaches, as well increases. This can come from rapid reproduction and as cases of bathers being injured and even killed. It growth during highly favorable environmental seems that human alterations to major ecosystems, conditions where asexual propagation exceeds physical which in turn have caused jelly outbreaks, are behind dilution. planktonic ecosystem shifts, and some anticipate that this will worsen in the future (Richardson et al., 2009). In situ evidence and anecdotic information on the Meanwhile, a recent review (Condon et al., 2012) salp outbreak formation brings some caution to these findings based on the little During the 2010 austral summer, several warnings of knowledge of the historical ecosystem baseline for salp outbreaks of the species Ihlea magalhanica were almost all jelly plankton species. The lack of historical recorded, mainly by salmon farmers, in the ISCh, background, the inaccuracy of the plankton nets used southern Chile. The first report occurred on 25 and sporadic sampling efforts is the main drawback to February, and lasted until 3 March 2010, when a solve the paradigm of the abundance of these groups. massive outbreak was recorded close to (Fig. In most cases, harmful gelatinous outbreaks are 1a).This bloom was the onset of the first alert which referred to as carnivorous zooplankton, mainly was followed by several observations in diverse areas cnidarians and occasionally ctenophores. Records of of the ISCh. During this outbreak, pictures were taken filter feeding zooplankton (salps, doliolids and by local salmon farmers and dissections of dead salmon pyrosomes) are less frequent and in most cases not even were performed in order to determine causes of ca. properly recorded. The latter group of zooplankton, due 25,000 salmon mortalities (Fig. 2a). Local salp to its nature, is harmless to humans and therefore not of abundances were not recorded, however farmer´s general concern. Salps have the ability to increase in estimate, indicate abundances of a couple of thousand number in short periods of time by the use of a complex per cubic meter. Massive concentrations clogged fish life cycle with an obligatory alternation of sexual and gills, probably the primary cause of fish deaths. asexual generations, allowing them to make rapid use Dissections of fish revealed the magnitude of this of resources when environmental conditions are outbreak as depicted by gut content analyses that sho- Massive salp outbreaks in the inner sea of Chiloé Island 606 3

Figure 1. a) Locations and dates where Ihlea maghalanica outbreaks were recorded along the coasts of the ISCh during 2010, 1) Quemchi (25 February-3 March), 2) Chaiten (10 March), 3) Maillen Island (20 March), 4) Cholgo Channel (24 March), 5) Caleta Arena (29 March), 6) Hueihue Bay (5 September), 7) Reloncaví (14 September -27 October), 8) Huelden (5 and 25-28 September). The color bar on the right hand side of (a) shows bathymetry (m), b) SeaWiFS true color image of whitish structure of 5 km in length by ca.1 km in width crossing the Cacahue Channel where an I. magalhanica outbreak was recorded during February 2010.

wed guts almost entirely filled with salps (Fig. 2b), Ihlea magalhanica is a subantarctic species with an which ultimately could cause fish starvation due to the oceanic distribution centered between the subtropical low energy content of salps compared to food pellets. convergence and the Antarctic convergence (Esnal & SeaWiFS true color images taken on 26 February Daponte, 1999). It is usually described as a steno- unveiled a whitish structure, 5 km in length by ca.1 km thermal cold water species (Ihle, 1958; Foxton, 1971), in width, crossing the Cacahue Channel, which was adapted to a temperature range between 4-16.8ºC most likely a result of this unusual outbreak (Fig. 1b). (Foxton, 1971; Daponte et al., 1993) while highest One week later, a second event was registered on March abundances are usually recoded in association to cold 10, close to Chaiten, 90 km away from the first report water masses (5-7ºC, Daponte et al., 1993). Few (Fig. 1a). After this, several warnings were received organisms were collected at the Magellan region during March across the ISCh: March 20 at Maillen (Michaelsen, 1907), south of 43.5ºS (F. Cabello, pers. Island (600 thousand oozooids and 54 thousand comm.) and off Valparaiso (Chile) (~33ºS, Fagetti, blastozooids per m-3), March 24 at Cholgo Channel and 1959), the later being the most northerly location ever March 29 at Caleta Arena (Fig. 2c). No more warnings recorded. of salp outbreaks were recorded until September in the Ihlea magalhanica, as most salp species, is usually north of Chiloé Island (Huelden Bay) and Sound found in large numbers in open ocean waters and (September to October 2010). exceptionally in neritic regions. Thus, the presence of 6074 Latin American Journal of Aquatic Research

of anomalous cold water masses into the ISCh), may favor the entrance and establishment of salp popu- lations, while local food availability conditions could

have allowed the massive increase in population abundance which were observed during 2010. The aim of this study was to explore the scenarios which have lead to the intrusion of I. magalhanica into the ISCh, as well as the environmental conditions which allowed the persistency and development of a

several massive outbreaks in this region during 2010.

MATERIALS AND METHODS

Sea surface temperature and surface current along the outer coast off Chiloé Island and the Patagonian fjord region Field data with a spatial resolution of 1/12º x 1/12º were obtained from the model outputs of Hybrid coordinate Ocean Model [HYCOM; Chassignet et al. (2009)] with

a Navy Coupled Ocean Data Assimilation (NCODA) System (Cummings, 2005) for data assimilation. The hybrid coordinate in HYCOM is isopycnal in the open, stratified ocean, but smoothly reverts to a terrain- following coordinate in shallow coastal regions, NCODA uses the model forecast as a first guess in a multivariate Optimal Interpolation scheme and assimilates available satellite altimeter observations satellite and in situ SST as well as available in situ vertical temperature and salinity profiles from XBTs, ARGO floats and moored buoys. Monthly means and their anomalies from January to March were calculated from daily sea surface temperature and the ocean surface velocity field (42º-46ºS, 74º-76ºW) from 2004 to 2011.

Figure 2. a) Salp abundance in a fish cage showing dead Chlorophyll-a and sea surface temperature in the salmons in the Maillen Island, b) gut content of dead ISCh salmon, c) salp abundance taken by scuba divers at Caleta In order to explore the general scenario in which I. Arena. magalhanica outbreaks were established and developed, we analyzed monthly means of historical satellite- derived chlorophyll-a (Chl-a) and sea surface this species in coastal waters is usually linked to the temperatures (4×4 km pixels, 11 µm night product) and horizontal advection of oceanic water masses into compared them with the monthly mean values for the coastal areas. The area of the east Pacific where salp year 2010. SST and Chl-a values were estimated from outbreaks were observed during 2010 coincides with the Moderate Resolution Imaging Spectro-radiometer the region in which the South Pacific Current (SPC) Aqua sensor (MODIS Aqua) from July 2002 to July reaches the eastern margin of the Pacific Ocean. Here, 2010. Data were averaged over an area which included the SPC splits into two branches; to the north, the all the locations where salps were seen (ca. 4000 km2) Humboldt Current or Chile-Peru Current and to the (Fig. 1a). Analysis used in this study was produced with south the Cape Horn Current. Forced by the SPC, the Giovanni online data system, developed and Subantarctic Water (SAAW) mass penetrates the inlets maintained by the NASA GES DISC (Acker & of the ISCh through the Guafo Passage, where it mixes Leptoukh, 2007). with the Estuarine Water (EW), forming the Subantarctic modal water (Silva et al., 1998; Sievers & Sampling and analysis Silva, 2006). We hypothesize that temporal anomalies On 5 September, during the first outbreak recorded in in the currents (i.e., displacement of the SPC, intrusion Huelden Bay (northern Chiloé Island), vertical Massive salp outbreaks in the inner sea of Chiloé Island 608 5

zooplankton samples (20-0 m) were collected on the (excluding posterior projections). Along with the third (and last) day of the outbreak with a WP-2 net zooplankton samples, vertical CTD casts (Seabird 19 (200 µm mesh size) equipped with a calibrated plus) were obtained during each sampling date from the flowmeter. Zooplankton samples were collected along surface down to 91 m. a transect consisting of five stations, 500 m apart, starting from close to the coast (ca. 100 m) into the Time series of phytoplankton functional group ISCh. After collection, the samples were fixed in 5% abundances prior to and after salp outbreaks buffered formaldehyde in sea water and stored for Phytoplankton abundance from January 2009 to further analysis. Abundance and identification of salp December 2011 was obtained for the three major basins species and life history phases were determined under along the ISCh: Reloncaví Sound, Ancud Sound, and a stereo microscope at 20x, by observing the entire Corcovado Gulf, and the region of the Desertores sample. Salps were identified using the taxonomic Islands. The amount of stations per basin varied from literature (Fagetti, 1959; Foxton, 1971; Esnal & between 2 (Reloncaví Sound) up to 52 (Desertores Daponte, 1999). Islands), while each basin was sampled between 1 and In order to explore the in situ potential effect of 24 times per month. Integrated samples (0-20 m) were salps on the autotrophic community, water samples for collected using a 2.5 cm diameter hose raised vertically determining phytoplankton and chlorophyll-a were and preserved in an acid Lugol’s solution. From these collected using 5-L PVC Go-Flo bottles at three depths samples, a 12 mL subsample was analyzed for diatom (0, 5 and 10 m). Subsamples (300 mL) were filtered and flagellate counting using standard methods through 0.7 µm GFF filters, immediately frozen (Utermöhl, 1958). (-20oC) for later pigment extraction and fluorometric analysis (Turner Design TD-700), according to Parsons RESULTS et al. (1984). To quantify phytoplankton abundance, 250 mL subsamples were fixed in acid Lugol’s solution Large scale offshore temperature and sea surface (1% final concentration) for phytoplankton cell counts. current climatology during austral summer A 50 mL aliquot was taken from each subsample and HYCOM model output derived SST climatology placed in a settling chamber for 30 h prior to analysis revealed the presence of cold SAAW water surrounding under an inverted microscope (Zeiss Axiovert 200, the outer coast from southern Chile during January 400x magnification) and counted using standard (Fig. 3a), this cold water approached the coast from the counting methods (Utermöhl, 1958). south-west around 45º30’S (Fig. 3d). Close to the coast, During the second salp outbreak at Huelden Bay the northward current deflected into two branches south which occurred between 25 and 28 September, salmon of the (~44ºS), one continued equatorward farm workers collected water samples using a 15 L around Guafo Island, while the second entered the ISCh -1 bucket at irregular intervals (0.5 to 10.5 h) at a fixed through the Guafo Passage at a speed of 0.06 m s (Fig. 3d). Equatorward currents flowed relatively slow close point within a salmon cage. The sampling point was -1 located on the edge of the salmon cage facing the inner to the coast (0.05 m s ) and tended to increase along a narrow band between 75º20’-75º40’W which expanded sea. Salps were sieved and directly counted (no north of the Guafo Island where it reached the coast of distinction between oozooids and blastozooids were Chiloé Island up to 75º4’W. made). Salp samples were fixed and sent to the laboratory for later taxonomic identification. During February, the currents ran in a predomi- nately meridional direction to the north with an increase Time series of salp abundance at the Reloncaví in velocity over a wider zonal range, while eastward currents along the western margin were less intense and Sound only apparent south of 45ºS (Fig. 3e). There was also Zooplankton samples were collected monthly by an increase in SST of ca. 1ºC associated with modifi- oblique tows using a Tucker trawl net (0.5 m-2 mouth cations in meridional currents over the entire study area area, 300 µm mesh size) equipped with a digital (Fig. 3b). The northern warm core tended to show a flowmeter for water volume quantification, during displacement to the south-east and low SST was only daytime from July to November 2010 at a fixed station seen in a restricted area close to the outer coast of the (41º32.6’S, 72º52’W). Samples were collected at 9 Patagonian fjord region and the Guafo Passage. depth intervals, from 5 to 100 m (0-5-10, 10-15, 15-20, During March, SST conditions were more similar to 20-30, 30-40, 40-50, 50-75 and 75-100 m) and those in January (Figs. 3a, 3c), with a cold SST in the preserved in borax buffered formalin (10%) for later outer fjord region, coupled with a change in the analysis. Salps were identified up to species level and meridional currents to a more oblique zonal current measured from the oral to the cloacal opening plane (eastward) during March (Fig. 3f). The oceanic 6096 Latin American Journal of Aquatic Research

Figure 3. Climatology of sea surface temperature (SST) for a) January, b) February, and c) March and climatology of sea surface currents (SSC) for d) January, e) February, f) March, along the outer coast off Chiloé Island. meridional currents were weaker and more homo- of cold water was projected to the ISCh, through the geneous (0.05-0.15 m s-1) with a slight increase in speed Guafo Passage. close to the continent (Fig. 3f). As in January, a branch Massive salp outbreaks in the inner sea of Chiloé Island 610 7

Large scale offshore SST anomalies and currents in a more restricted range. During January, SST and prior to and during outbreaks in austral summer satellite derived Chl-a were close to their historical 2010 mean (14.1ºC and 10 mg m-3, respectively). However, During January 2010 the oceanic region off Chiloé during February the basin experienced a significant Island and the fjord region experienced a cooling in cooling of 1ºC in relation to its historical mean and relation to its climatology (Figs. 4a, 5a). South of the 0.7ºC below standard error (Fig. 5a). This drop was not Guafo Passage a decrease of ca. 0.5ºC was observed, followed by any significant change in Chl-a while positive anomalies were mainly located in the concentration, which increased slightly during this northern region close to the coastal environment. Sea month. After this event, temperatures rose again to surface currents during this month displayed a wider “normal” conditions while chlorophyll experienced a predominance of shoreward zonal currents (Fig. 4d), substantial drop from normal values of 13.5 to 4 mg -3 which deflected to the north as they approached the Chl-a m (Fig. 5b). After this drop, Chl-a did not continent. Meridional currents along outer recover and remained low throughout the entire year. displayed a similar spatial trend, while slowing down in Unfortunately during June, satellite images could not relation to its climatology. During February, the entire be obtained due to cloud cover (Fig. 5b). area exhibited a decrease in SST (Fig. 4b) and sea surface currents (Fig. 4e). Offshore negative SST Time series of phytoplankton functional groups in anomalies reached as low as -1.5ºC, covering almost the four major regions in the ISCh the entire south western area. The rest of the area Throughout the annual cycle, the three basins and experienced an SST drop of 0.5 and 1ºC. Concomitant Desertores Islands were dominated by chain forming to the decrease in SST a significant decrease in offshore diatoms such as Chaetoceros spp., Skeletonema spp. current was observed. Meridional surface currents and Thalassiosira spp. In the austral spring and summer -1 decreased to less than 0.05 m s in the oceanic region, (September to May), these species tend to form blooms, while currents close to the coast and interior of the especially in the three largest regions (Ancud Sound, Guafo Passage did not change significantly with respect Desertores Islands and Corcovado Gulf) while the to the climatology (Figs. 4b, 4e). Zonal currents were Reloncaví Sound has a less marked seasonality with weak, always heading north, similar to climatologically sporadic autumn and winter blooms (Fig. 6). conditions. During March, conditions tended to Phytoplankton cell concentrations show a north to stabilize; SST showed a slight decrease of 0.5ºC over south gradient with higher concentrations in the the entire area, while currents remained close to that Reloncaví and Ancud sounds, declining in the which can be expected for this month (Figs. 4c, 4f). Desertores Islands. This region functions as a barrier between the highly productive northern Ancud Sound General scenario in the ISCh (Ancud Sound) prior to and during salp outbreaks in summer 2010 and the southern Corcovado Gulf, where the lowest phytoplankton abundances were measured. On an annual time scale, temperatures displayed a typical seasonality with maximum values during austral A marked reduction in phytoplankton spring bloom summer (14-13ºC, January to March) and a continuous in 2010 was evident in the Ancud Sound, Desertores decline from autumn to winter, with minimum SST Islands and Corcovado Gulf, in which phytoplankton during July and August (9.8ºC) (Fig. 5a). Subsequently, abundance decreased down to half of what was later temperatures started to rise during spring through to recorded in 2011, with no evident change in species summer. Surface Chl-a however presented a more composition between years. From May through variable cycle during the year and is partially decoupled September 2010, there was a marked reduction in with temperature (Fig. 5b). High concentrations of Chl- abundance of large phytoplankton cells at Reloncaví a were typically observed in austral autumn rather than Sound, during which small cell numbers of the genus in summer. During January and February (austral Chaetoceros and other less representative algae’s summer), Chl-a concentrations fluctuated around 10 dominated. During 2009 the Ancud Sound experienced mg Chl-a m-3, March and April being the months with an unusual extension for ca. two months of this less the highest satellite Chl-a concentrations (~13.5 mg productive phase due to the early decline of the summer Chl-a m-3). Usually after May there is a substantial drop bloom and coincided with a drop in cell abundance of one order in magnitude in Chl-a (10.5 to 0.9 mg Chl- which lasted until the end of 2010. The Ancud Sound a m-3) when it reaches its lowest value along the annual experienced a massive bloom of small flagellates cycle (Fig. 5b). Tetraselmis spp., during the end of spring 2009 and During 2010, SST in the Ancud Sound experienced beginning of summer 2010, just after the last recorded the similar annual drop as that seen offshore, although outbreak of I. magalhanica in 2010. 6118 Latin American Journal of Aquatic Research

Figure 4. Mean monthly sea surface temperature anomalies (SST) during a) January, b) February and c) March 2010, and sea surface currents (SSC) recorded during d) January, e) February and f) March 2010 along the outer coast off Chiloé Island.

Ihlea magalhanica sampling and analysis the austral summer by local salmon farmers at the same Huelden (Ancud Sound) location almost three weeks later. The total abundance Zooplankton samples taken during the first outbreaks in of aggregate and solitary forms reached up to 1.13 ind Huelden (5 September 2010), were low in salp m-3 (Fig. 7a) and were dominated by aggregate abundance compared to the outbreaks reported during individuals (93% of the total population) as expected. Massive salp outbreaks in the inner sea of Chiloé Island 612 9

Figure 5. a) Monthly mean climatology (black dots) and standard error (grey area) of SST, compared with monthly mean SST (white dots) recorded during 2010 in Ancud Sound, b) monthly mean climatology (black dots) and standard error (grey area) of satellite Chl-a, compared with monthly mean satellite Chl-a (white dots) recorded during 2010 in Ancud Sound. Black arrows indicate months in which Ihlea magalhanica outbreaks were recorded. Vertical bars denote number of pixel counts used to compute mean values of SST and Chl-a in monthly averages during 2010.

The size of the aggregate phase was of 7.7 ± 5.3 mm in 7b).The main contributors (over 98.7%) to flagellate length (mean ±; n = 79) while small buds (~1 mm abundance were small sized (<20 µm) loricated length) which were not considered for the abundance flagellates, followed by small (<20 µm) naked counts, represented 31% of total measured individuals. dinoflagellates (0.5%), Gonyaulax spp. (0.4%), Solitary forms were only collected at stations 1 and 2, Diplopsalis spp. (0.3%) and others (0.1%). Diatoms, with a mean size of 22 ± 4 mm (mean ± SD; n = 6). The coccolitophores and ciliates did not show any apparent total abundance was higher at the two stations located change in their relative abundances that could be close to the coast with 0.9-1.13 ind m-3. These stations associated with changes in salp abundance (Fig. 7b). experienced a sharp decline of Chl-a (Fig. 7a) Twenty days after this event and in this same associated with the decline of flagellates, which were location, a massive outbreak of I. magalhanica took the dominant microplankton group in all stations with place (Fig. 7b). Abundance on this occasion was around over 58% of the total abundance (mean 64%) (Fig. 10,000 ind m-3 in the first 3 h after the first measure- 10613 Latin American Journal of Aquatic Research

Figure 6. Average (2009-2011) of monthly time series of abundance of the dominant phytoplankton (cells mL-1) of the upper 20 m water column along the four major basins in the ISCh. Black bars above each graph indicate the number of samples collected per month in each basin, used to calculate the monthly mean values while arrows show months in which Ihlea magalhanica outbreaks were recorded.

ments were taken. Afterwards, their abundance increa- decreasing approximately 4-fold in a range of just 3 h. sed steadily over the next 21 h, reaching a maximum of These temporal oscillations in abundance were coupled 33,330 and 32,000 ind m-3 at 21:30 and 24:00 on 25 with the tidal phase. The highest abundances usually September 2010, respectively. This drastic increase is recorded in periods of tidal rise, while decreases were most likely a result of aggregation rather than observed during the ebbing tide (Fig. 7c). The reproduction taking into account the rapid increase in zooplankton samples were taken from a salmon cage just a couple of hours. Consecutive changes in facing towards the inner sea, and therefore, observed abundance were of great magnitude, increasing/ increases are mainly a result of the aggregation and clo- Massive salp outbreaks in the inner sea of Chiloé Island 614 11

Figure 7. a) Ihlea magalhanica abundance (ind m-3) collected at Huelden stations during 5 September 2010, and chlorophyll concentration (mg Chl-a m-3), 2.5 km perpendicular to the coast transect, b) abundance of main groups of phytoplankton (cells m-3) along the off shore transect, c) tidal height during massive I. magalhanica outbreak in Huelden bay between 25 and 28 September (source http://www.shoa.cl), d) I. magalhanica abundance (ind m-3) recorded by salmon farmers between 25 and 28 September. Grey areas indicate night time samplings.

gging of salps in salmon farm nets during the rising discard changes in abundance due to vertical tide. While during the ebbing tide, salps were migrations. transported towards to the inner sea, reducing abundance significantly close to the farm. This Reloncaví Sound outbreak lasted for almost three days after the first The I. magalhanica abundances recorded between July observations, reducing gradually on the afternoon of and November 2010 show a clear outbreak event in the the third day. By the fourth day no noticeable Reloncaví Sound. While from July to August no abundances were recorded by the salmon farmers, individuals were collected. During September the first therefore sampling was terminated. Potential appearance of salps were observed with a large relationships between the abundance of organisms and predominance of adult and young solitary forms, which the diel cycle were not observed which lead us to are both close to the abundance of aggregate individuals 12615 Latin American Journal of Aquatic Research

Figure 8. Abundance and vertical distribution of Ihlea magalhanica reproductive life stages at the Reloncaví Sound during: a) September and b) October 2010. Vertical distribution of temperature: c) and salinity, d) during both sampling periods.

(aggregate/solitary mean ratio of 2.1) (Fig. 8a). Salps September. In parallel with the observed changes in showed low abundance at surface (<0.23 ind m-3 in the salp abundance, the hydrographic characteristics at the 0-10 m depth stratum), and increased sharply up to 18.5 Reloncaví Sound also changed significantly from ind m-3 below a depth of 10 m where salps were September to October 2010 (Figs. 8c, 8d). During distributed almost evenly down to a depth of 50 m (Fig. September, a vertical quasi-homogeneous thermal 8a). Their abundance declined again sharply to less than distribution (10.4ºC) was found (Fig. 8c) with a weak 0.4 ind m-3 below 50 m. One and a half months later saline gradient, forming a fragile halocline at a depth of (mid November), a new massive outbreak took place 25 m (Fig. 8c). In October, a strong vertical thermal and (up to 628 ind m-3, Fig. 8b), in which a change to their haline gradient was formed with low salinity (30.6) and relative reproductive life cycle phase contribution a high temperature (12.7ºC) at surface, decreasing occurred as depicted by the decrease in the relative sharply at 20 m where the base of the mixed layer is abundance of solitary forms to levels similar to those formed (Figs. 8c, 8d). previously recorded in Huelden, and the aggregate phase dominated overall (aggregate/solitary mean ratio DISCUSSION of 18.8). Vertical distribution was uneven with low abundance The Patagonian fjord offshore region is influenced by on the surface and a steady increase from 200 ind m-3 the eastward flow of the SPC, which reaches the coast at the 5-10 m stratum up to 422 ind m-3 at the 15-20 m of South America at ~45ºS were it deflects, forming the depth stratum (Fig. 8b). Under this depth range, there Cape Horn Current to the south, and the Humboldt was a region (20-40 m depth) that showed a decline in Current to the north. Part of the SAAW near the coast abundance and then an abrupt increase in the 40-50 m enters at subsurface depths (see below) to the ISCh stratum, coinciding with the higher abundance recorded through a narrow 2 km, 150 m deep channel (Chacao at this station (628 ind m-3). Finally, below this stratum Channel, 42ºS) to the north of Chiloé Island, and the abundances decreased down to 0.2 ind m-3 from 50 to Guafo Passage (43.5ºS), a 37 km wide channel with an 100 m depth; following the same pattern observed in entrance at sill ca. 150 m depth (Silva et al., 1995). SAAW Massive salp outbreaks in the inner sea of Chiloé Island 616 13

develops isolating the lower SAAW, nutrient-rich (nitrate and orthophosphate) water (Silva et al., 1998; González et al., 2010). The particular geographical structure of the ISCh, i.e., being split into three sub- basins, allows for the development of high numbers of large chain-forming diatom blooms which (see Fig. 6), are not only limited to the austral spring and summer seasons but are also frequent in winter and autumn. (2) The central sub-region (Ancud Sound) is a larger basin with less intense haline stratification and a seasonal production coupled with the light regime, supporting a rich phytoplankton biomass, largely dominated by large chain-forming diatoms (Fig. 6, González et al., 2010). (3) The southernmost basin (Corcovado Gulf) is separated from the middle basin (Ancud Sound) through a string of islands (Desertores Islands). The Corcovado Gulf connects with the Pacific Ocean through the Guafo Passage, where the SAAW mixes with fresh waters coming from rivers, generating a lower salinity water mass (31 to 33, Silva et al., 1998). The entire area between Desertores Islands and up to

the Corcovado Channel is less productive than the Figure 9. Salmon farm concessions granted until September 2009. (Source: Subsecretaria de Pesca, 2009). enclosed northern and central basins, while a clear seasonality is still observed (Fig. 6, Lara et al., 2010).

Proliferation of salps were only recorded in the which enters through the Guafo Passage has a low Reloncaví and Ancud sounds, north of Desertores salinity signature (~33, Palma & Silva, 2004), derived Islands, while no records were registered in the from the northward costal transport of fresh water Corcovado Gulf, despite several salmon farms existing discharges along the fractured coast line from 47 to in this region that might have documented the event 43ºS (Dávila et al., 2002). This low salinity water mass, (ca. 170 fish farms, Fig. 9). Why the salp outbreaks results from the mix of SAAW and Estuarine Water developed exclusively in the northern innermost basins (EW), which forms the Modified Subantarctic Water is not yet clear. One possible hypothesis is that the (MSAAW). The general circulation pattern through the Desertores Islands may function as a natural barrier, Guafo Passage is an offshore outflow of MSAAW at creating a retention area for planktonic organisms, by the surface (upper 25 m) and an inflow of SAAW at limiting the circulation and dispersal of organisms depth (25-110 m) (Palma & Silva, 2004). Thus the across the central and southern basins while providing entrance of subantarctic species into the ISCh under a suitable area for aggregation and development, as normal conditions should occur at the subsurface. sugested by Tello & Rodriguez‐Benito (2009). Salp However, during some episodic events such as those outbreaks, such as those observed in this study, occur observed during January 2010, a change in the explosively, facilitated by their direct development and circulation pattern associated with the intensification of release of progeny at relatively large sizes, high growth zonal currents can cause a massive entrance of SAAW rates and short generation times as the reported for into the ISCh, as revealed by a temperature drop in the other salp species such as Thalia democratica (Madin Guafo Passage and around the ISCh from January to & Purcell, 1992; CIESM, 2001). A drastic increase in February 2010 (Fig. 4e). This intrusion was most abundance was observed in September 2010 at the probably responsible for the transport of oceanic station located at the Ancud Sound, where tens of subantarctic species, such as I. magalhanica into this solitary forms were collected initially, while at the end coastal semi enclosed habitat. of October, six weeks later, abundances reached up to The ISCh is geographically divided into three 600 ind m-3, mainly dominated by aggregates, clearly defined sub-basins (Tello & Rodriguez‐Benito, indicating that this species is most likely capable to 2009): (1) the northern one being the Reloncaví Sound, complete its life cycle in the ISCh. The absence of a semi enclosed system influenced by a strong silicic- salps, during the months before or after the outbreaks, acid-rich freshwater input and strong stratification in are not necessarily a result of a lack of reproduction the upper 4-7 m. Beneath this layer, a strong nutricline over the entire area but most likely related to the patchy 14617 Latin American Journal of Aquatic Research

The reason why outbreaks did not developed in the past is unknown, despite of the presence of salps in this region in previous years and that intensive aquaculture established since the 1990´s. Prior recorded abundances of salps in the ISCh were usually low, probably limiting a successful reproduction, which leads to assume that the intrusion of SAAW might have transported an already developed outbreak into the ISCh. The potential effect on the phytoplankton community in the ISCh might be considerable when taking into account that they filter continuously throughout the day and also that each salp may display very high filtering rates as in the case of Pegea confoederata that can exert grazing pressure equivalent to that of several hundreds of large calanoid copepods

(Harbison & Gilmer, 1976) considering that the filtration rate is not known for I. magalhanica. This is especially relevant given the high concentrations -3 recorded of up to 654,000 salps m close to coastal net pens and 600 ind m-3 in the more open systems in the Reloncaví Sound.

Accordingly, the effect of the salp outbreaks on the ISCh pelagic food web might be considerable taking into account the drastic decrease in phytoplankton cell abundance and Chl-a biomass. If this decrease was mainly result of grazing we also might expect a substantial export of organic carbon out of the photic Figure 10. Inner Sea of Chiloé locations where salp zone via fecal pellets. Salps are able to filter on a wide abundance was detected. Information was compiled from size range of particles above 0.1 µm, which results in samplings carried out during CIMAR Cruises 9, 12 and an aggregation of small particles into large and fast 13. Filled circles show areas where salps were recorded, sinking fecal pellets. Sinking rates of salp fecal pellets while open circles represent areas where salps were range from between 59 up to 2238 m d-1 (Bruland & absent. Numbers in the legends indicate salp density (ind Silver, 1981; Madin, 1982), which implies an extre- m-3). mely efficient export of organic matter and nutrients from the upper water column down to the benthos. distribution of this event. The fast increase and in most Massive sinking events of fecal pellets and cases short persistency of the outbreaks in a particular senescent bodies during salp outbreaks have been area makes it extremely difficult to follow and almost previously reported in other regions (Bathmann, 1988). impossible to predict. Thus, most of the warnings we On that occasion, salps with an abundance more than received could not be properly investigated and upon 700 ind m-3 were able to deplete surface nutrients by our arrival (one to two days after warnings) the salps exporting organic matter and nutrients out of the photic had already gone. zone off the coast of Ireland. This process could explain Reports of salps in Patagonia are scarce and usually why phytoplankton was not able to recover its biomass include numbers less than 0.052 ind. m-3 (Fig. 10). over several months after the first outbreaks in the These have only been recorded during winter time and ISCh. The only phytoplankton species which showed have been absent during spring and summer. Thus, the an unusual increase were small flagellates, which are outbreaks during 2010 are the highest recorded for this capable of migrating through the water column to the group of species to date and occurred in summer and nutricline and have lower nutrient requirements than spring. The first large-scale outbreaks were recorded in diatoms because of their larger surface to volume ratio. late February 2010, a situation in which the ISCh On the other hand, the decrease in cell abundances of offshore and inshore temperature showed strong large, chain-forming diatoms (especially those equipped negative anomalies (-1.5oC) and a month later with long setae and protuberances) throughout 2010 accompanied by Chl-a concentrations well below the would also favor the development of I. magalhanica historical averages expected for this region (Fig. 5b). growth, since large phytoplankton have shown to be Massive salp outbreaks in the inner sea of Chiloé Island 618 15

deleterious to filtering activity, causing clogging of hand has being linked to climate-driven variability feeding structures (Harbison et al., 1986). (Condon et al., 2012) and to the proliferation of Gelatinous species with short generation times such settlements and increased human activities in coastal as ctenophores (Sullivan et al., 2001) siphonophores areas (Purcell, 2012). The inner sea of Chiloé is not an (Pagès et al., 2001) and pelagic tunicates (Purcell & exception, where proliferations of the scyphomedusa Madin, 1991), have the ability to make use of Chrysaora plocamia and Phacellophora catschatica aggregation for reproductive purposes to maintain have affected salmon farming installations (Palma et and/or increase their population density. In this study, al., 2007; Bravo et al., 2011). Unfortunately, and as field observations indicated that salmon net pens can same as most cases, reports as same as impacts are effectively act as a physical barrier to salp movement, mainly based on anecdotic information given by local promoting the clogging, increasing local salp aggre- farmers (see Mianzan et al., 2014), and more reliable gations. Thanks to short generation times, these information based on scientific orientated approaches massive aggregations along salmon net pens have the are extremely scarce. The random nature of the potential to increase reproductive encounter rates, occurrence (spatial and temporal) basically associated which can result in a highly effective instrument to to a complex life cycle and “r” strategy (Boero et al., maximize reproductive success and promote outbreak 2008), makes gelatinous zooplankton outbreaks very formation (Heron & Benham, 1985). hard to follow. Aggregations have generally been described by the Future presence of geological barriers (i.e., shoreline and seafloor) and hydrographic (i.e., horizontal fronts and There is no clear evidence of which kind of remote vertical discontinuities) (reviewed by Graham et al., forcing might have been involved during the intrusion 2001) or mesoscale eddies (Everett et al., 2011). of oceanic water into the ISCh, and whether it might However, the effect of man-made barriers is less well increase/decrease in frequency in the future. There is documented (Purcell et al., 2007; Doyle et al., 2008; however some hints from remote atmospheric Baxter et al., 2011; Mianzan et al., 2014). The ISCh anomalies that occurred during the end of 2009 and the possess the majority of the country’s salmon farms with beginning of 2010 which might have been partially over 500 fish farms along the ISCh coast (excluding responsible for the build-up of this scenario which fjords) (Fig. 10), thus the potential use of these favors the intrusion of SAAW into the northern ISCh. structures as aggregation barriers to achieve a high In the last few decades it has become evident that reproductive success is enormous. climate variability in the southern hemisphere is mainly forced by the variability on timescales from intra- Effect on aquaculture seasonal to interannual by the Southern Annular Mode Warnings of high densities of salps close to salmon (SAM), which is characterized by a large scale farms occurred seven times in 2010, while adverse alteration of atmospheric mass between mid and high- effects were only recorded on one occasion. High salp latitudes; modifying the latitudinal distribution of the density inside fish cages hamper normal fish feeding on westerly winds. Positive SAM is associated with a food pellets, as observed in the guts of dead salmon significant increase in SST in the Subtropical Zone collected from cages. The stomachs of these fish were (STZ) (Lovenduski & Gruber, 2005), while inverse entirely filled with salps, while gills were clogged by patterns occur during negative anomalies (Hall & salp remains. Several salmon biopsies could not reveal Visbeck, 2002). the exact cause of death; however, veterinarians Positive anomalies usually persist throughout all suggested that gill damage and the reduction of the seasons, but are strongest during the austral summer oxygen diffusion due to clogging might be the primary (Marshall, 2003). During the austral summer 2010, an cause of death. unusual negative SAM was documented (Fig. 11) Overall, Chl-a levels and cell density remained low coinciding with strong negative SST anomalies at the throughout the entire year, which not only favored the outer Patagonian region. If this anomaly triggered the persistence of salps (until the end of October) but also entrance of I. magalhanica to the ISCh, two processes showed serious negative consequences for blue mussel must have occurred simultaneously to achieve these farmers, who experienced decreased growth of adults outbreaks. First we would need a substantial transport and little settlement of juveniles in the following season of SAAW towards to the coast, which ultimately must (http://www.mundoacuicola.cl/comun/?modulo-=3& carry enough organisms to facilitate a successful view=1&cat=1&idnews=341). Deleterious effect of reproduction to generate an abundant offspring. The gelatinous zooplankton on aquaculture has becoming achievement of these two conditions makes these increased attention in the last decades; which on one scenario relatively improbable phenomena to occur on 16619 Latin American Journal of Aquatic Research

Bathmann, U.V. 1988. Mass occurrence of Salpa fusiformis in the spring of 1984 off Ireland: implications for sedimentation processes. Mar. Biol., 97(1): 127-135. Baxter, E.J., M.M. Sturt, N.M. Ruane, T.K. Doyle, R. McAllen, L. Harman & H.D. Rodger. 2011. Gill damage to Atlantic salmon (Salmo salar) caused by the common jellyfish (Aurelia aurita) under experimental challenge. PloS ONE, 6: 1-6. Boero, F., J. Bouillon, C. Gravili, M.P. Miglietta, T. Parsons & S. Piraino. 2008. Gelatinous plankton: irregularities rule the world (sometimes). Mar. Ecol. Prog. Ser., 356: 299-310. Bravo, V., S. Palma & N. Silva. 2011. Seasonal and Figure 11. Marshal re-analysis of the Southern Annular vertical distributional patterns of medusae in Aysén Mode (SAM), time series between January 2007 and region, southern Chile. Lat. Am. J. Aquat. Res., 39(2): August 2010. Grey area indicates the month where first 359-377. outbreaks and negative temperature anomalies were Bruland, K.W. & M.W. Silver. 1981. Sinking rates of recorded in the fjord region. Data obtained from fecal pellets from gelatinous zooplankton (Salps, http://www.nerc-bas.ac.uk/icd/gjma/sam.html. Pteropods, Doliolids). Mar. Biol., 63(3): 295-300. Chassignet, E.P., H.E. Hurlburt, E.J. Metzger, O.M. a regular basis. The exact mechanism controlling the Smedstad, J.A. Cummings, G.R. Halliwell, R. Bleck, coupling between climate forcing and zooplankton R. Baraille, A.J. Wallcraft, C. Lozano, H.L. Tolman, outbreaks needs further study and many questions A.Srinivasan, S. Hankin, P. Cornillon, R. Weisberg, A. remain unanswered. Examining the role of the Barth, R. He, F. Werner & J. Wilkin. 2009. US meridional circulation across the Southern Pacific and GODAE Global Ocean Prediction with the HYbrid the advection of oceanic water masses into the ISCh Coordinate Ocean Model (HYCOM). Oceanography, and its relation to remote forcing such as the SAM 22(2): 64-75. variability might help to unveil some of these answers. Condon, R.H., W.M. Graham, C.M. Duarte, K.A. Pitt, Additional modeling work, both regionally and C.H. Lucas, S.H.D. Haddock, K.R. Sutherland, K.L. globally, can help to demonstrate biological and Robinson, M.N. Dawson, M.B. Decker, C.E. Mills, physical coupling and allow potential predictions on J.E. Purcell, A. Malej, H. Mianzan, S.I. Uye, S. changes in productivity and help to anticipate potential Gelcich & L.P. Madin. 2012. Questioning the rise of invasions. gelatinous zooplankton in the World's Oceans. BioScience, 62(2): 160-169. ACKNOWLEDGEMENTS Cummings, J.A. 2005. Operational multivariate ocean data assimilation. Q.J.R. Meteorol Soc., 131(613): This research was supported by the COPAS-Sur Austral 3583-3604. project and a FONDECYT 1100534 grant to Leonardo Daponte, M.C., F.L. Capitanio, L. Machinandiarena & Castro. We are grateful to Rodrigo Cristi and Dr. Dirk G.B. Esnal. 1993. Planktonic tunicates (Chordata, Schories for providing information and photographs of outbreaks. We also thank Marine Harvest for providing Tunicata) of the RTMA "Evrika" in the southwestern technical assistance during sampling campaigns and L. Atlantic Ocean. Iheringia, 74: 71-80. Cisternas and M.J. Cuevas for sap sorting from Dávila, P.M., D. Figueroa & E. Müller. 2002. Freshwater zooplankton samples. To Dr. Héctor Paves and Program input into the coastal ocean and its relation with the CIMAR-Fiordos 9, 12 and 13 (CONA-SHOA) to provide salinity distribution off austral Chile (35-55°S). Cont. data on abundance of salps collected on several Shelf Res., 22: 521-534. oceanographic expeditions along the fjord region of the Deibel, D. 1982a. Laboratory-measured grazing and southern Chilean Patagonia. ingestion rates of the salp, Thalia democratica Forskal, and the doliolid, Dolioletta gegenbauri Uljanin REFERENCES (Tunicata, Thaliacea). J. Plankton Res., 4(2): 189-201. Deibel, D. 1982b. Laboratory determined mortality, Acker, J.G. & G. Leptoukh. 2007. Online analysis fecundity and growth rates of Thalia democratica enhances use of NASA Earth science data. EOS. Forskal and Dolioletta gegenbauri Uljanin (Tunicata, Trans. Am. Geophys. Union, 88(2): 14-17. Thaliacea). J. Plankton Res., 4(1): 143-153. Massive salp outbreaks in the inner sea of Chiloé Island 620 17

Doyle, T.K., H. De Haas, D. Cotton, B. Dorschel, V. Heron, A.C. & E.E. Benham. 1985. Life history Cummins, J.D. Houghton, J. Davenport & G.C. Hays. parameters as indicators of growth rate in three salp 2008. Widespread occurrence of the jellyfish Pelagia populations. J. Plankton Res., 7(3): 365-379. noctiluca in Irish coastal and shelf waters. J. Plankton Ihle, J.E.W. 1958. Salpidae. In: H.G. Bronns (ed.). Res., 30(8): 963-968. Klassen und Ordnungen des Tierreichs Akademische Esnal, G. & M.C. Daponte. 1999. Salpida. In: D. Verlagsgesellschaft Geest und Portig K.G., Leipzig, Boltovskoy (ed.). South Atlantic zooplankton. pp. 241-401. Backhuys, Leiden, pp. 1423-1444. Javidpour, J., U. Sommer & T. Shiganova. 2006. First Everett, J.D., M.E. Baird & I.M. Suthers. 2011. Three- record of Mnemiopsis leidyi A. Agassiz 1865 in the dimensional structure of a swarm of the salp Thalia Baltic Sea. Aquat Invasions, 1(4): 299-302. democratica within a cold-core eddy off southeast Kideys, A.E. 1994. Recent dramatic changes in the Black Australia. J. Geophys. Res., 116: C12046. Sea ecosystem: the reason for the sharp decline in Faasse, M.A. & K.M. Bayha. 2006. The ctenophore Turkish anchovy fisheries. J. Mar. Syst., 5(2): 171- Mnemiopsis leidyi (Agassiz, 1865) in coastal waters of 181. the Netherlands: an unrecognized invasion? Aquat. Kideys, A.E. 2002. Fall and rise of the Black Sea Invasions, 1(4): 270- 277. ecosystem. Science, 297(5586): 1482-1484. Fagetti, E. 1959. Salpas colectadas frente a las costas Lara, C., M. Miranda, V. Montecino & J.L. Iriarte. 2010. central y norte de Chile. Rev. Biol. Mar. Valparaiso, Chlorophyll-a MODIS mesoscale variability in the 9(1-3): 201-228. Inner Sea of Chiloé, Patagonia, Chile (41-43°S): Foxton, P. 1971. On Ihlea magalhanica (Apstein, 1894) Patches and gradients? Rev. Biol. Mar. Oceanogr., (Tunicata, Salpidae) and lhlea racovitzai (Van 45(2): 217-225. Beneden). Discovery Rep., 35: 179-198. Le Borgne, R. & P. Moll. 1986. Growth rates of the salp González, H.E., M.J. Calderón, L. Castro, A. Clement, Thalia democratica in Tikehau atoll (Tuamotuis.). L.A. Cuevas, G. Daneri, J.L. Iriarte, L. Lizárraga, R. Oceanogr. Trop., 21(1): 23-29. Martínnez, E. Menschel, N. Silva, C. Carrasco, C. Lovenduski, N.S. & N. Gruber. 2005. Impact of the Valenzuela, C.A. Vargas, & C.Molinet. 2010. Primary southern annular mode on southern ocean circulation production and plankton dynamics in the Reloncaví and biology. Geophys. Res. Lett., 32: L11603. Fjord and the Interior Sea of Chiloé, Northern Patagonia, Chile. Mar. Ecol. Prog. Ser., 402: 13-30. Madin, L.P. 1982. Production, composition and sedimentation of salp fecal pellets in oceanic waters. Goy, J., P. Morand & M. Etienne. 1989. Long-term Mar. Biol., 67(1): 39-45. fluctuations of Pelagia noctiluca (Cnidaria, Scypho- medusa) in the western Mediterranean Sea. Prediction Madin, L.P. & J.E. Purcell. 1992. Feeding, metabolism, by climatic variables. Deep Sea Res. (I Oceanogr. Res. and growth of Cyclosalpa bakeri in the subarctic Pap.), 36(2): 269-279. Pacific. Limnol. Oceanogr., 37(6): 1236-1251. Graham, W.M., F. Pagès & W.M. Hamner. 2001. A Marshall, G.J. 2003. Trends in the Southern Annular physical context for gelatinous zooplankton Mode from observations and reanalyses. J. Climate, aggregations: a review. Hydrobiologia, 451(1): 199- 16(24): 4134-4143. 212. Mianzan, H., J. Quiñones, S. Palma, A. Schiariti, E.M. Hall, A. & M. Visbeck. 2002. Synchronous variability in Acha, K.L. Robinson & W.M. Graham. 2014. the southern hemisphere atmosphere, sea ice, and Chrysaora plocamia: a poorly understood jellyfish ocean resulting from the annular mode. J. Climate, from South American waters. In: K.A. Pitt & C.H. 15(21): 3043-3057. Lucas (eds.). Jellyfish Blooms. Springer Netherlands, pp. 219-236. Harbison, G.R. & R.W. Gilmer. 1976. The feeding rates of the pelagic tunicate Pegea confoederata and two Michaelsen, W. 1907. Tunikaten. Friedrichsen & Co, other salps. Limnol. Oceanogr., 21(4): 517-528. Hamburg, 1:1-84. Harbison, G.R., V.L. McAlister & R.W. Gilmer. 1986. Molinero, J.C., F. Ibanez, P. Nival, E. Buecher & S. The response of the salp, Pegea confoederata, to high Souissi. 2005. North Atlantic climate and levels of particulate material: starvation in the midst of northwestern Mediterranean plankton variability. plenty. Limnol. Oceanogr., 31(2): 371-382. Limnol. Oceanogr., 50(4): 1213-1220. Heron, A.C. 1972. Population ecology of a colonizing Mullin, M.M. 1983. In situ measurement of filtering rates species: the pelagic tunicate Thalia democratica. of the salp, Thalia democratica, on phytoplankton and Oecologia, 10(4): 269-293. bacteria. J. Plankton Res., 5(2): 279-288. 62118 Latin American Journal of Aquatic Research

Pagès, F., H.E. González, M. Ramón, M. Sobarzo & J.-M. Sievers, H. & N. Silva 2006. Masas de agua y circulación Gili. 2001. Gelatinous zooplankton assemblages en los canales y fiordos australes. In: N. Silva & S. associated with water masses in the Humboldt Current Palma (eds.). Avances en el conocimiento oceano- System, and potential predatory impact by Bassia gráfico de las aguas interiores chilenas, Puerto Montt bassensis (Siphonophora: Calycophorae). Mar. Ecol. a Cabo de Hornos. Comité Oceanográfico Nacional- Prog. Ser., 210: 13-24. Pontificia Universidad Católica de Valparaíso, Valparaíso, pp. 53-58. Palma, S. & N. Silva. 2004. Distribution of siphonophores, chaetognaths, euphausiids and oceanographic condi- Silva, N., H.A. Sievers & R. Prado. 1995. Características tions in the fjords and channels of southern Chile. oceanográficas y una proposición de circulación para Deep-Sea Res. Part II, 51(6-9): 513-535. algunos canales australes de Chile entre 41º20’S y 46º40’S. Rev. Biol. Mar., Valparaíso, 30(2): 207-254. Palma, S., P. Apablaza & N. Silva. 2007. Hydromedusae (Cnidaria) of the Chilean Southern channels (from the Silva, N., C. Calvete & H.A. Sievers. 1998. Masas de agua Corcovado Gulf to the Pulluche-Chacabuco channels). y circulación general para algunos canales australes Sci. Mar., 71(1), 65-74. entre Puerto Montt y Laguna San Rafael, Chile (Crucero Cimar-Fiordo 1). Cienc. Tecnol. Mar., 21: Parsons, T.R, R. Maita & C.M. Lalli. 1984. Counting, 17-48. media and preservatives. A manual of chemical and biological methods for seawater analysis. Pergamon Sullivan, B.K., D.V. Keuren & M. Clancy 2001. Timing Press, Oxford, UK, 157-164. and size of blooms of the ctenophore Mnemiopsis leidyi in relation to temperature in Narragansett Bay, Purcell, J.E. 2012. Jellyfish and ctenophore blooms RI. Hydrobiologia, 451(1): 113-120. coincide with human proliferations and environmental perturbations. Annu Rev. Mar. Sci., 4(1): 209-235. Sutherland, K.R., L.P. Madin & R. Stocker. 2010. Filtration of submicrometer particles by pelagic Purcell, J.E. & L.P. Madin. 1991. Diel patterns of tunicates. Proc. Nat. Acad. Sci., 107(34): 15129- migration, feeding, and spawning by salps in the 15134. subarctic Pacific. Mar. Ecol. Prog. Ser., 73: 211-217. Tello, G.A. & C. Rodriguez‐Benito. 2009. Characte- Purcell, J.E., S.I. Uye & W.T. Lo. 2007. Anthropogenic rization of mesoscale spatio‐temporal patterns and causes of jellyfish blooms and their direct variability of remotely sensed Chl-a and SST in the consequences for humans: a review. Mar. Ecol. Prog. Interior Sea of Chiloé (41.4-43.5ºS). Int. J. Remote Ser., 350: 153-174. Sens., 30(6): 1521-1536. Richardson, A., A. Bakun, G. Hays & M. Gibbons 2009. The Mediterranean Science Commission (CIESM). 2001. The jellyfish joyride: causes, consequences and Gelatinous zooplankton outbreaks: theory and practice, management responses to a more gelatinous future. Naples, 29 August -1st September 2001. CIESM, Trends Ecol. Evol., 24: 312-334. Monaco, 110 pp. Shiganova, T.A., Z.A. Mirzoyan, E.A Studenikina, S.P. Utermöhl, H. 1958. Zur Vervollkommnung der quanti- Volovik, I. Siokou-Frangou, S. Zervoudaki, E.D. tativen Phytoplankton-Methodik. Mitt. Int. Ver. Christou, A.Y. Skirta & H.J. Dumont. 2001. Theor. Angew. Limnol., 9: 1-39. Population development of the invader ctenophore Mnemiopsis leidyi , in the Black Sea and in other seas Uye, S.I. 2008. Blooms of the giant jellyfish Nemopilema of the Mediterranean basin. Mar. Biol., 139(3): 431- nomurai: a threat to the fisheries sustainability of the 445. East Asian Marginal Seas. Plank. Benth. Res., 3(Suppl.): 125-131.

Received: 15 October 2013; Accepted: 20 June 2014

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.